Funnelling: A New Ensemble Method for Heterogeneous Transfer Learning and Its Application to Cross-Lingual Text Classification.
Andrea EsuliAlejandro MoreoFabrizio SebastianiPublished in: ACM Trans. Inf. Syst. (2019)
Keyphrases
- cross lingual
- text classification
- transfer learning
- ensemble methods
- labeled data
- prediction accuracy
- text categorization
- machine learning methods
- machine learning
- learning tasks
- benchmark datasets
- decision trees
- feature selection
- random forest
- text mining
- bag of words
- text data
- naive bayes
- sentiment classification
- text documents
- cross domain
- unlabeled data
- n gram
- sentiment analysis
- knn
- language modeling
- multi task
- multi label
- text classifiers
- feature subset
- k nearest neighbor
- knowledge representation
- active learning
- information retrieval